Drafted through my n8n + AI pipeline, edited by me.
Most automation projects start in the wrong place, on the flashy thing instead of the frequent one. Here is the test I use to decide what to automate first, a quick way to check any task, and the jobs worth removing before anything else.
The test for what to automate first
Good first candidates pass three checks. Miss any one and the project gets harder than it should be.
High-frequency: it happens many times a week, not once a quarter.
Rule-clear: the decision is a rule a new hire could follow, not a judgment call.
Low-judgment: being slightly wrong is cheap and easy to catch.
A decision tree: frequent, rule-based, low-risk tasks get automated first; judgment work stays human; rare tasks wait; broken processes get fixed first.
Does this task happen most days?
Walk a task through this before you automate it.
What I'd automate first, in order
Data moving between tools by hand.
The same follow-up after the same trigger.
Routing and tagging incoming requests.
Chasing a status someone has to remember to check.
What to leave alone
A process that is still broken. Fix it by hand first.
Judgment work: pricing a non-standard deal, handling an upset customer.
Anything rare enough that maintaining the automation costs more than it saves.
Automate the work that is boring and certain. Keep the work that is rare and consequential. Most teams do the opposite and wonder why it all feels fragile.
Bring me the workflow that eats your week. I'll tell you what I'd automate first, and what I'd leave human.